The risk of cancer associated with most missense mutations in the BRCA1 and BRCA2 genes cannot be determined. Thus, when a missense mutation is identified in the BRCA1 or BRCA2 breast and ovarian cancer predisposition genes during clinical testing, the patient with the mutation and the patient's family members cannot benefit from improved risk assessment. The difficulty in interpreting whether the 1,600 known missense mutations in BRCA1 and BRCA2 predispose to cancer stems from the undefined influence of these mutations on BRCA1 and BRCA2 function. The hypotheses to be addressed in this proposal is that missense mutations in the BRCA1 and BRCA2 genes can be defined as either cancer predisposing or of no clinical significance using a combination of genetic, epidemiological and molecular biological approaches. Missense mutations in the BRCA1 BRCT domains and the BRCA2 DNA binding domain will be characterized using (A) a likelihood model of cancer causality for missense mutations based on co- segregation of the mutations with cancer, co-occurrence of the mutations with other deleterious mutations, and on the family history profile of cancer in families carrying these mutations;(B) functional assays;(C) evolutionary sequence conservation analysis;and (D) a two-component mixture model that utilizes the data generated in A) - C). Each method will be used to evaluate missense mutations from these domains and to independently predict which mutations are deleterious or neutral. These data will also be combined in the two-component mixture model to generate a combined estimate of the cancer relevance of each missense mutation and to identify the assays/approaches that are most valuable for classifying missense mutations. Importantly, once the sensitivity and specificity of the functional assays and sequence conservation method have been determined for some of the more common mutations in conjunction with the genetic/family data it will then be possible to use these assays to define the cancer relevance of other rare missense mutations in these domains of BRCA1 and BRCA2, even in the absence of genetic data. At the conclusion of the study the cancer relevance of many missense mutations will have been determined and provided to the carriers who will then be able to benefit from improved cancer risk assessment, cancer prevention and perhaps even therapeutic strategies